(Revised) DESCRIPTION (adapted from the applicant's abstract): HIV/AIDS rates remain high in young adults. Although college students are aware of how HIV/AIDS is transmitted, many continue to engage in risky behaviors. Three studies are proposed to examine the role of alcohol in college students' decisions about engaging in risky behavior. Laboratory studies are proposed so that participants can be randomly assigned to their beverage condition. Prior to the laboratory session, participants complete a survey that includes measures of their own behavior and attitudes and experiences related to likelihood of engaging in risky behavior when intoxicated (e.g., usual alcohol consumption, expectancies, and attitudes). Each laboratory study includes three independent variables: 1) gender of the participant, 2) participants' drink condition (Intoxicated, Placebo, Sober), and 3) degree of risk associated with various behaviors. By including a placebo cell, expectancy and pharmacological effects of alcohol can be compared. Measures of cognitive impairment and anxiety will be included to examine the processes hypothesized to mediate the relationship between intoxication and willingness to engage in risky behavior. Degree of risk in the situation is operationalized differently in each of the three studies. Each of the studies also uses different stimulus materials to enhance the generalizability of the findings. Hypotheses will be tested about gender differences in the types of situations that present the most conflict about risky behavior. These studies' findings can be used to develop prevention programs tailored to address factors that encourage young women and men to engage in risky behaviors.